Data on the petroglyphs of the Eastern Desert of Egypt that have recently become available are collated and analysed in detail. Images of wild animals, domestic animals, anthropoids and boats, together with geometric patterns, are classified and analysed by statistical means to reach conclusions about the preferences of the artists in terms of subject matter, style, context and geographical distribution. Published data on the petroglyphs of the Nile Valley are analysed similarly but in less detail to permit comparison with the Eastern desert, identifýing many similarities but also significant differences. Petroglyphs from farther afield - the Western Desert, Uweinat and the Gilf Kebir, south-eastern Egypt, Sinai, the Negev and Arabia - are also compared. The general conclusion is that similarity decreases rapidly with distance, but there are also a few individual cases indicating that some of the artists were in contact No evidence of contact with Arabia is found. Dating of the Eastern Desert petroglyphs, both relative and absolute, by various methods, is reviewed. It is concluded that, while precision is not possible, it is reasonably certain that many of the animal images, in particular those of animals that require a relatively moist environment, and some of the boats and anthropoids, date to the fourth millennium BC, while other boats, anthropoids and desert animals are later. The significance of the petroglyphs to the artists who drew them is addressed. Various possible reasons for drawing them, such as religious, magical or shamanistic practices, are examined and it is concluded that in the majority of cases it is not possible to say what they meant, Only in the case of the boats is it possible to be reasonably sure that they had some sort of fimerary significance. The probability that many of the images had little or no significance to the artists is recognised. Nevertheless it is possible to deduce much about the artists and their communities. A series of propositions, each based on the analysis of the corpus of images and without any assumptions about meaning, is drawn up. Together these propositions form a partial description of the society that gave rise to the petroglyphs, although not its beliefs. This description is objective and free from conjecture.